Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease that affects 30,000 Americans each year. Of these 30,000 Americans, it has been suggested that up to 50% will experience cognitive and behavioral changes in the form of frontotemporal dysfunction and up to 40% will meet criteria for frontotemporal dementia (FTD). Riluzole the only FDA approved agent for ALS extends a patient's lifespan by 2-3 months, and there are no proven therapies for the cognitive changes associated with ALS. More effective therapy for this universally fatal disease is needed. Results from an open label pilot trial of 20 patients treated with memantine at 10 mg twice a day (BID) suggested that treatment with the combination of memantine and riluzole slowed ALS disease progression. This trial also showed that levels of specific protein biomarkers in the cerebrospinal fluid (CSF) at baseline correlated with the rate of disease progression. A concurrent Phase 2 study found no effect with similar dosing; however, the study was limited in terms of power. Comments on previous failed drug trials in ALS have raised the concern that many ALS trials study a potential therapeutic agent at only a single dose and thus may miss the potential efficacy of non FDA approved doses; therefore, this proposed study will test a higher dose of memantine, 20 mg BlD, in a double blind, placebo controlled, randomized trial of 90 patients with ALS to determine if a combination therapy of memantine with riluzole can slow disease progression compared to treatment with riluzole alone. The primary outcome measure will be the rate of disease progression as measured by the ALS Functional Rating Scale-Revised (ALSFRS-R). ln addition the investigators will examine the cognitive deficits seen in ALS patients measured by the ALS Cognitive Behavioral Screen (ALS-CBS) and the Neuropsychiatric lnventory Questionnaire (NlP-Q). Finally, specific validated protein biomarkers found in the CSF will be examined to determine if there is a correlation between the levels of these biomarkers and the rate of disease progression. ln particular the investigators will measure the ratio of phosphorylated heavy neurofilament to complement 3 (pNFH/C3) to see if this ratio is predictive of disease progression and if the levels change during therapy with memantine.